MRUNet-3D: A multi-stride residual 3D UNet for lung nodule segmentation.

Obtaining an accurate segmentation of the pulmonary nodules in computed tomography (CT) images is challenging. This is due to: (1) the heterogeneous nature of the lung nodules; (2) comparable visual characteristics between the nodules and their surroundings. A robust multi-scale feature extraction mechanism that can effectively obtain multi-scale representations at a granular level can improve segmentation accuracy. As the most commonly used network in lung nodule segmentation, UNet, its variants, and other image segmentation methods lack this robust feature extraction mechanism. In this study, we propose a multi-stride residual 3D UNet (MRUNet-3D) to improve the segmentation accuracy of lung nodules in CT images. It incorporates a multi-slide Res2Net block (MSR), which replaces the simple sequence of convolution layers in each encoder stage to effectively extract multi-scale features at a granular level from different receptive fields and resolutions while conserving the strengths of 3D UNet. The proposed method has been extensively evaluated on the publicly available LUNA16 dataset. Experimental results show that it achieves competitive segmentation performance with an average dice similarity coefficient of 83.47 % and an average surface distance of 0.35 mm on the dataset. More notably, our method has proven to be robust to the heterogeneity of lung nodules. It has also proven to perform better at segmenting small lung nodules. Ablation studies have shown that the proposed MSR and RFIA modules are fundamental to improving the performance of the proposed model.

Study protocol for Near-infrared molecular imaging for lung cancer detection and treatment during mini-invasive surgery (phase II Trial) - (the RECOGNISE study).

INTRODUCTION: To date, radical surgery remains the best curative option in patients with early-stage lung cancer. In patients with small lung lesions, video-assisted thoracic surgery (VATS) should be increasingly chosen as a fundamental alternative to thoracotomy as it is associated with less postoperative pain and better quality of life. This scenario necessarily increases the need for thoracic surgeons to implement new localization techniques. The conventional near-infrared (NIR) indocyanine green (ICG) method demonstrated a significant limitation in deep cancer recognition, principally due to its intrinsic low-depth tissue penetration. Similarly, the lymph-node sentinel approach conducted by the ICG method was demonstrated to be inefficient, mainly due to the non-specificity of the tracker and the irregular pathway of pulmonary lymph node drainage. Our study aims to evaluate the effectiveness of Cetuximab- IRDye800CW in marking lung nodules and mediastinal lymph nodes. METHODS AND ANALYSIS: This study is defined as an open-label, single-arm, single-stage phase II trial evaluating the effectiveness of Cetuximab-IRDye800CW in detecting tumors and lymph-node metastases in patients with lung cancer who are undergoing video-assisted thoracic surgery (VATS). Cetuximab is a monoclonal antibody that binds, inhibits, and degrade the EGFR. The IRDye® 800CW, an indocyanine-type NIR fluorophore, demonstrated enhanced tissue penetration compared to other NIR dyes. The combination with the clinical approved monoclonal antibody anti-epidermal growth factor EGFR Cetuximab (Cetuximab-IRDye800) has shown promising results as a specific tracker in different cancer types (i.e., brain, pancreas, head, and neck). The study's primary outcome is focused on the proportion of patients with lung nodules detected during surgery using an NIR camera. The secondary outcomes include a broad spectrum of items, including the proportion of patients with detection of unexpected cancer localization during surgery by NIR camera and the proportion of patients with negative surgical margins, the evaluation of the time spawns between the insertion of the NIR camera and the visualization of the nodule and the possible morbidity of the drug assessed during and after the drug infusion. ETHICS AND DISSEMINATION: This trial has been approved by the Ethical Committee of Azienda Ospedaliera Universitaria Città della Salute e della Scienza di Torino (Torino, Italy) and by the Italian Medicines Agency (AIFA). Findings will be written as methodology papers for conference presentations and published in peer-reviewed journals. The Azienda Ospedaliera Universitaria Città della Salute e della Scienza di Torino, the University of Torino, and the AIRC Public Engagement Divisions will help identify how best to publicize the findings.Trial registration EudraCT 202,100,645,430. CLINICALTRIALS: gov NCT06101394 (October 23, 2023).

First-in-human use of a new robotic electromagnetic navigation bronchoscopic platform with integrated Tool-in-Lesion Tomosynthesis (TiLT) technology for peripheral pulmonary lesions: The FRONTIER study.

BACKGROUND AND OBJECTIVE: As the presentation of pulmonary nodules increases, the importance of a safe and accurate method of sampling peripheral pulmonary nodules is highlighted. First-generation robotic bronchoscopy has successfully assisted navigation and improved peripheral reach during bronchoscopy. Integrating tool-in-lesion tomosynthesis (TiLT) may further improve yield. METHODS: We performed a first-in-human clinical trial of a new robotic electromagnetic navigation bronchoscopy system with integrated digital tomosynthesis technology (Galaxy System, Noah Medical). Patients with moderate-risk peripheral pulmonary nodules were enrolled in the study. Robotic bronchoscopy was performed using electromagnetic navigation with TiLT-assisted lesion guidance. Non-specific results were followed up until either a clear diagnosis was achieved or repeat radiology at 6 months demonstrated stability. RESULTS: Eighteen patients (19 nodules) were enrolled. The average lesion size was 20 mm, and the average distance from the pleura was 11.6 mm. The target was successfully reached in 100% of nodules, and the biopsy tool was visualized inside the target lesion in all cases. A confirmed specific diagnosis was achieved in 17 nodules, 13 of which were malignant. In one patient, radiological monitoring confirmed a true non-malignant result. This translates to a yield of 89.5% (strict) to 94.7% (intermediate). Complications included one pneumothorax requiring observation only and another requiring an overnight chest drain. There was one case of severe pneumonia following the procedure. CONCLUSION: In this first-in-human study, second-generation robotic bronchoscopy using electromagnetic navigation combined with integrated digital tomosynthesis was feasible with an acceptable safety profile and demonstrated a high diagnostic yield for small peripheral lung nodules.

Ultrathin Bronchoscopy Without Virtual Navigation for Diagnosis of Peripheral Lung Lesions.

BACKGROUND: The increasing incidence of encountering lung nodules necessitates an ongoing search for improved diagnostic procedures. Various bronchoscopic technologies have been introduced or are in development, but further studies are needed to define a method that fits best in clinical practice and health care systems. RESEARCH QUESTION: How do basic bronchoscopic tools including a combination of thin (outer diameter 4.2 mm) and ultrathin bronchoscopes (outer diameter 3.0 mm), radial endobronchial ultrasound (rEBUS) and fluoroscopy perform in peripheral pulmonary lesion diagnosis? STUDY DESIGN AND METHODS: This is a retrospective review of the performance of peripheral bronchoscopy using thin and ultrathin bronchoscopy with rEBUS and 2D fluoroscopy without a navigational system for evaluating peripheral lung lesions in a single academic medical center from 11/2015 to 1/2021. We used a strict definition for diagnostic yield and assessed the impact of different variables on diagnostic yield, specifically after employment of the ultrathin bronchoscope. Logistic regression models were employed to assess the independent associations of the most impactful variables. RESULTS: A total of 322 patients were included in this study. The median of the long axis diameter was 2.2 cm and the median distance of the center of the lesion from the visceral pleural surface was 1.9 cm. Overall diagnostic yield was 81.3% after employment of the ultrathin bronchoscope, with more detection of concentric rEBUS views (93% vs. 78%, p < 0.001). Sensitivity for detecting malignancy also increased from 60.5% to 74.7% (p = 0.033) after incorporating the ultrathin scope into practice, while bronchus sign and peripheral location of the lesion were not found to affect diagnostic yield. Concentric rEBUS view, solid appearance, upper/middle lobe location and larger size of the nodules were found to be independent predictors of successful achievement of diagnosis at bronchoscopy. INTERPRETATION: This study demonstrates a high diagnostic yield of biopsy of lung lesions achieved by utilization of thin and ultrathin bronchoscopes. Direct visualization of small peripheral airways with simultaneous rEBUS confirmation increased localization rate of small lesions in a conventional bronchoscopy setting without virtual navigational planning.

Research Priorities for Interventions to Address Health Disparities in Lung Nodule Management: An Official American Thoracic Society Research Statement.

Background: Lung nodules are common incidental findings, and timely evaluation is critical to ensure diagnosis of localized-stage and potentially curable lung cancers. Rates of guideline-concordant lung nodule evaluation are low, and the risk of delayed evaluation is higher for minoritized groups. Objectives: To summarize the existing evidence, identify knowledge gaps, and prioritize research questions related to interventions to reduce disparities in lung nodule evaluation. Methods: A multidisciplinary committee was convened to review the evidence and identify key knowledge gaps in four domains: 1) research methodology, 2) patient-level interventions, 3) clinician-level interventions, and 4) health system-level interventions. A modified Delphi approach was used to identify research priorities. Results: Key knowledge gaps included 1) a lack of standardized approaches to identify factors associated with lung nodule management disparities, 2) limited data evaluating the role of social determinants of health on disparities in lung nodule management, 3) a lack of certainty regarding the optimal strategy to improve patient-clinician communication and information transmission and/or retention, and 4) a paucity of information on the impact of patient navigators and culturally trained multidisciplinary teams. Conclusions: This statement outlines a research agenda intended to stimulate high-impact studies of interventions to mitigate disparities in lung nodule evaluation. Research questions were prioritized around the following domains: 1) need for methodologic guidelines for conducting research related to disparities in nodule management, 2) evaluating how social determinants of health influence lung nodule evaluation, 3) studying approaches to improve patient-clinician communication, and 4) evaluating the utility of patient navigators and culturally enriched multidisciplinary teams to reduce disparities.

Combined model integrating deep learning, radiomics, and clinical data to classify lung nodules at chest CT.

OBJECTIVES: The study aimed to develop a combined model that integrates deep learning (DL), radiomics, and clinical data to classify lung nodules into benign or malignant categories, and to further classify lung nodules into different pathological subtypes and Lung Imaging Reporting and Data System (Lung-RADS) scores. MATERIALS AND METHODS: The proposed model was trained, validated, and tested using three datasets: one public dataset, the Lung Nodule Analysis 2016 (LUNA16) Grand challenge dataset (n = 1004), and two private datasets, the Lung Nodule Received Operation (LNOP) dataset (n = 1027) and the Lung Nodule in Health Examination (LNHE) dataset (n = 1525). The proposed model used a stacked ensemble model by employing a machine learning (ML) approach with an AutoGluon-Tabular classifier. The input variables were modified 3D convolutional neural network (CNN) features, radiomics features, and clinical features. Three classification tasks were performed: Task 1: Classification of lung nodules into benign or malignant in the LUNA16 dataset; Task 2: Classification of lung nodules into different pathological subtypes; and Task 3: Classification of Lung-RADS score. Classification performance was determined based on accuracy, recall, precision, and F1-score. Ten-fold cross-validation was applied to each task. RESULTS: The proposed model achieved high accuracy in classifying lung nodules into benign or malignant categories in LUNA 16 with an accuracy of 92.8%, as well as in classifying lung nodules into different pathological subtypes with an F1-score of 75.5% and Lung-RADS scores with an F1-score of 80.4%. CONCLUSION: Our proposed model provides an accurate classification of lung nodules based on the benign/malignant, different pathological subtypes, and Lung-RADS system.

Promoting lung cancer screening of high-risk patients by primary care providers.

BACKGROUND: Lung cancer screening (LCS) with low-dose computed tomography (LDCT) of the chest of eligible patients remains low. Accordingly, augmentation of appropriate LCS referrals by primary care providers (PCPs) was sought. METHODS: The quality improvement (QI) project was performed between April 2021 and June 2022. It incorporated patient education, shared decision-making (SDM) with PCPs, and tracking of initial LDCT completion. In each case, lag time (LT) to LCS and pack-years (PYs) were calculated from initial LCS eligibility. The cohort's scores were compared to national scores. Patient zip codes were used to create a geographic map of our cohort for comparison with public health data. RESULTS: An immediate and sustained increase in weekly LCS referrals from PCPs was recorded. Of 337 initial referrals, 95% were men, consisting of 66.2% Black, 28.4% White, and 5.4% other. Mean PY was less for minorities (45.3 vs. 37.3 years; p = .0002) but mean LT was greater for Whites (7.9 vs. 6.2 years; p = .03). Twenty-five percent of veterans failed to report to their scheduled screening, and two declined referrals. Notably, most no-show patients lived in transit deserts. Furthermore, Lung-RADS scores 4B/4X were more than double the expected prevalence (p = .008). CONCLUSIONS: The PCPs in this study successfully augmented LCS referrals. A substantial proportion of these patients were no-shows, and our data suggest complex racial and socioeconomic factors as contributing variables. In addition, a higher-than-expected number of initial Lung-RADS scores 4B/4X were reported. A large, multisite QI project is warranted to address overcoming potential transportation barriers in high-risk patient populations.

TP53 mutation prevalence in normal airway epithelium as a biomarker for lung cancer risk.

BACKGROUND: There is a need for biomarkers that improve accuracy compared with current demographic risk indices to detect individuals at the highest lung cancer risk. Improved risk determination will enable more effective lung cancer screening and better stratification of lung nodules into high or low-risk category. We previously reported discovery of a biomarker for lung cancer risk characterized by increased prevalence of TP53 somatic mutations in airway epithelial cells (AEC). Here we present results from a validation study in an independent retrospective case-control cohort. METHODS: Targeted next generation sequencing was used to identify mutations within three TP53 exons spanning 193 base pairs in AEC genomic DNA. RESULTS: TP53 mutation prevalence was associated with cancer status (P < 0.001). The lung cancer detection receiver operator characteristic (ROC) area under the curve (AUC) for the TP53 biomarker was 0.845 (95% confidence limits 0.749-0.942). In contrast, TP53 mutation prevalence was not significantly associated with age or smoking pack-years. The combination of TP53 mutation prevalence with PLCOM2012 risk score had an ROC AUC of 0.916 (0.846-0.986) and this was significantly higher than that for either factor alone (P < 0.03). CONCLUSIONS: These results support the validity of the TP53 mutation prevalence biomarker and justify taking additional steps to assess this biomarker in AEC specimens from a prospective cohort and in matched nasal brushing specimens as a potential non-invasive surrogate specimen.

Robotics in the diagnosis and staging of lung cancer.

The diagnosis of peripheral small lung lesions by electromagnetic navigational bronchoscopy is still inferior to computed tomography (CT) guided percutaneous transthoracic needle lung biopsy. Robotic bronchoscopy is a new technology that may be a potential breakthrough in the diagnosis of small lung lesions. Real-time tools such as electromagnetic navigation, radial-endobronchial ultrasound, and cone beam CT may further improve the diagnostic yield rate may further improve the diagnostic yield rate. In this article, we reviewed early experience of robotic bronchoscopy for diagnosis and staging of lung cancer.

Robotic bronchoscopic needle-based confocal laser endomicroscopy to diagnose peripheral lung nodules.

BACKGROUND AND OBJECTIVE: Robotic bronchoscopy has demonstrated high navigational success in small peripheral lung nodules but the diagnostic yield is discrepantly lower. Needle based confocal laser endomicroscopy (nCLE) enables real-time microscopic imaging at the needle tip. We aim to assess feasibility, safety and needle repositioning based on real-time nCLE-guidance during robotic bronchoscopy in small peripheral lung nodules. METHODS: Patients with suspected peripheral lung cancer underwent fluoroscopy and radial EBUS assisted robotic bronchoscopy. After radial EBUS nodule identification, nCLE-imaging of the target area was performed. nCLE-malignancy and airway/lung parenchyma criteria were used to identify the optimal sampling location. In case airway was visualized, repositioning of the biopsy needle was performed. After nCLE tool-in-nodule confirmation, needle passes and biopsies were performed at the same location. MEASUREMENTS AND MAIN RESULTS: Twenty patients were included (final diagnosis n = 17 (lung) cancer) with a median lung nodule size of 14.5 mm (range 8-28 mm). No complications occurred. In 19/20 patients, good quality nCLE-videos were obtained. In 9 patients (45%), real-time nCLE-imaging revealed inadequate positioning of the needle and repositioning was performed. After repositioning, nCLE-imaging provided tool-in-nodule-confirmation in 19/20 patients. Subsequent ROSE demonstrated representative material in 9/20 patients (45%) and overall diagnostic yield was 80% (16/20). Of the three patients with malignant nCLE-imaging but inadequate pathology, two were diagnosed with malignancy during follow-up. CONCLUSION: Robotic bronchoscopic nCLE-imaging is feasible and safe. nCLE-imaging in small, difficult-to-access lung nodules provided additional real-time feedback on the correct needle positioning with the potential to optimize the sampling location and diagnostic yield.

Diagnostic Yield of Shape-Sensing Robotic-Assisted Bronchoscopy for Ground-Glass Nodules and Subsolid Nodules with a Solid Component Less than 6 mm.

BACKGROUND: Ground-glass pulmonary nodules (GGNs) are most commonly sampled by percutaneous transthoracic biopsy. Diagnostic yield for ground-glass nodules using robotic-assisted bronchoscopy has been scarcely described, with a reported yield of 70.6%. OBJECTIVES: The aim of this study is to assess diagnostic yield for GGNs using shape-sensing robotic-assisted bronchoscopy (ssRAB). METHOD: A retrospective study of patients who underwent ssRAB for evaluation of GGNs, from September 2021 to April 2023. Primary outcome was diagnostic yield of ssRAB for GGNs, secondary outcomes were sensitivity for malignancy, and complications that required admission or intervention. RESULTS: A total of 23 nodules were biopsied from 22 patients. Median age was 71 years (IQR 66-81), 63.6% were female, and 40.9% had a previous history of cancer. Forty-three percent of nodules were in the right upper lobes, and the median lesion size was 1.8 × 1.21. Twelve were subsolid nodules (SSNs), and 11 were pure GGNs. Overall diagnostic yield was 87%, with a sensitivity for malignancy of 88.9%. Adenocarcinoma was the most common malignancy diagnosed (70%). No procedure-related complications were reported. CONCLUSION: The use of ssRAB shows a high diagnostic yield for diagnosing GGN and SSN with less than 6 mm solid component with a low risk for complications.

Solitary Lung Nodule: CT-Guided Transthoracic Biopsy vs Transbronchial Biopsy With Endobronchial Ultrasound and Flexible Bronchoscope, a Meta-Analysis of Randomized Controlled Trials.

BACKGROUND: Transbronchial lung biopsy with radial endobronchial ultrasound (rEBUS-TBB) and Computed tomography (CT) scan-guided transthoracic biopsy (CT-TTB) are commonly used to investigate peripheral lung nodules but high-quality data are still not clear about the diagnostic and safety profile comparison of these two modalities. METHOD: We included all randomized controlled trials (RCT) comparing rEBUS-TBB with a flexible bronchoscope and CT-TTB for solitary lung nodules. Two reviewers extracted data independently on diagnostic performance and complication rates. RESULTS: 170 studies were screened, 4 RCT with a total of 325 patients were included. CT-TTB had a higher diagnostic yield than rEBUS-TBB (83.45% vs 68.82%, risk difference - 0.15, 95% CI, [- 0.24, - 0.05]), especially for lesion size 1-2 cm (83% vs 50%, risk difference - 0.33, 95% CI, [- 0.51, - 0.14]). For malignant diseases, rEBUS-TBB had a diagnostic yield of 75.75% vs 87.7% of CT-TTB. rEBUS-TBB had a significant better safety profile with lower risks of pneumothorax (2.87% vs 21.43%, OR = 0.12, 95% CI [0.05-0.32]) and combined outcomes of hospital admission, hemorrhage, and pneumothorax (8.62% vs 31.81%, OR 0.21, 95% CI, [0.11-0.40]). Factors increasing diagnostic yield of rEBUS were lesion size and localization of the probe but not the distance to the chest wall and hilum. CONCLUSION: CT-TTB had a higher diagnostic yield than rEBUS-TBB in diagnosing peripheral lung nodules, particularly for lesions from 1 to 2 cm. However, rEBUS-TBB was significantly safer with five to eight times less risk of pneumothorax and composite complications of hospital admission, hemorrhage, and pneumothorax. The results of this study only apply to flexible bronchoscopy with radial ebus without navigational technologies. More data are needed for a comparison between CT-TTB with rEBUS-TBB combined with advanced navigational modalities.

Assessing the impact of socioeconomic status on incidental lung nodules at an urban safety net hospital.

INTRODUCTION: Lower socioeconomic status has been identified as an emerging risk factor for health disparities, including lung cancer outcomes. Most research investigating these outcomes includes patients from formal lung cancer screening programs. There is a paucity of studies assessing the relationship between socioeconomic status and incidental lung nodules. This study aimed to investigate the association between socioeconomic status and the size of incidental lung nodules on initial presentation at an urban safety net hospital, which did not have a formal lung cancer screening program or incidental lung nodule program. METHODS: A retrospective chart review was conducted on patients with incidental lung nodules on CT chest imaging who were referred from primary care to a pulmonology clinic at a safety net hospital. Patients with incomplete nodule characteristics information were excluded. Data on demographics, comorbidities, smoking history, insurance type, immigration status, and geographical factors were collected. Less commonly studied determinants such as crime index, cost of living, and air quality index were also assessed. Logistic regression analysis was performed to assess relationships between nodule size and socioeconomic determinants. RESULTS: Out of 3,490 patients with chest CT scans, 268 patients with ILNs were included in the study. 84.7% of patients represented racial or ethnic minorities, and most patients (67.8%) had federal insurance. Patients with non-commercial insurance were more likely to have larger, inherently higher-risk nodules (> 8 mm) compared to those with commercial insurance (OR 2.18, p 0.01). Patients from areas with higher unemployment rates were also less likely (OR 0.75, p 0.04) to have smaller nodules (< 6 mm). Patients representing racial or ethnic minorities were also more likely to have nodules > 8 mm (OR 1.6, p 0.24), and less likely to have nodules < 6 mm (OR 0.6, p 0.32), however, these relationships were not statistically significant. CONCLUSION: This study found that lower socioeconomic status, indicated by having non-commercial insurance, was associated with larger incidental lung nodule size on initial presentation. While it is established that socioeconomic status is associated with disparities in lung cancer screening, these findings suggest that inequalities may also be present in those with incidental lung nodules. Further research is needed to understand the underlying mechanisms and develop interventions to address these disparities in incidental lung nodule evaluation and improve outcomes.

A novel fusion algorithm for benign-malignant lung nodule classification on CT images.

The accurate recognition of malignant lung nodules on CT images is critical in lung cancer screening, which can offer patients the best chance of cure and significant reductions in mortality from lung cancer. Convolutional Neural Network (CNN) has been proven as a powerful method in medical image analysis. Radiomics which is believed to be of interest based on expert opinion can describe high-throughput extraction from CT images. Graph Convolutional Network explores the global context and makes the inference on both graph node features and relational structures. In this paper, we propose a novel fusion algorithm, RGD, for benign-malignant lung nodule classification by incorporating Radiomics study and Graph learning into the multiple Deep CNNs to form a more complete and distinctive feature representation, and ensemble the predictions for robust decision-making. The proposed method was conducted on the publicly available LIDC-IDRI dataset in a 10-fold cross-validation experiment and it obtained an average accuracy of 93.25%, a sensitivity of 89.22%, a specificity of 95.82%, precision of 92.46%, F1 Score of 0.9114 and AUC of 0.9629. Experimental results illustrate that the RGD model achieves superior performance compared with the state-of-the-art methods. Moreover, the effectiveness of the fusion strategy has been confirmed by extensive ablation studies. In the future, the proposed model which performs well on the pulmonary nodule classification on CT images will be applied to increase confidence in the clinical diagnosis of lung cancer.

DEHA-Net: A Dual-Encoder-Based Hard Attention Network with an Adaptive ROI Mechanism for Lung Nodule Segmentation.

Measuring pulmonary nodules accurately can help the early diagnosis of lung cancer, which can increase the survival rate among patients. Numerous techniques for lung nodule segmentation have been developed; however, most of them either rely on the 3D volumetric region of interest (VOI) input by radiologists or use the 2D fixed region of interest (ROI) for all the slices of computed tomography (CT) scan. These methods only consider the presence of nodules within the given VOI, which limits the networks' ability to detect nodules outside the VOI and can also encompass unnecessary structures in the VOI, leading to potentially inaccurate segmentation. In this work, we propose a novel approach for 3D lung nodule segmentation that utilizes the 2D region of interest (ROI) inputted from a radiologist or computer-aided detection (CADe) system. Concretely, we developed a two-stage lung nodule segmentation technique. Firstly, we designed a dual-encoder-based hard attention network (DEHA-Net) in which the full axial slice of thoracic computed tomography (CT) scan, along with an ROI mask, were considered as input to segment the lung nodule in the given slice. The output of DEHA-Net, the segmentation mask of the lung nodule, was inputted to the adaptive region of interest (A-ROI) algorithm to automatically generate the ROI masks for the surrounding slices, which eliminated the need for any further inputs from radiologists. After extracting the segmentation along the axial axis, at the second stage, we further investigated the lung nodule along sagittal and coronal views by employing DEHA-Net. All the estimated masks were inputted into the consensus module to obtain the final volumetric segmentation of the nodule. The proposed scheme was rigorously evaluated on the lung image database consortium and image database resource initiative (LIDC/IDRI) dataset, and an extensive analysis of the results was performed. The quantitative analysis showed that the proposed method not only improved the existing state-of-the-art methods in terms of dice score but also showed significant robustness against different types, shapes, and dimensions of the lung nodules. The proposed framework achieved the average dice score, sensitivity, and positive predictive value of 87.91%, 90.84%, and 89.56%, respectively.

An Improved Convolution Neural Network and Modified Regularized K-Means-Based Automatic Lung Nodule Detection and Classification.

If lung cancer is not detected in its initial phases, it can be fatal. However, because of the quantity and structure of its nodules, lung cancer is difficult to detect early. For accurate detections, radiologists require assistance from automated tools. Numerous expert methods have been created over time to assist radiologists in the diagnosis of lung cancer. However, this requires accurate research. Therefore, in this article, we propose a framework to precisely detect lung cancer by categorizing it between benign and malignant nodules. To achieve this objective, an efficient deep-learning algorithm is presented. The presented technique consists of four stages, namely pre-processing, segmentation, classification, and severity stage analysis. Initially, the collected image is given to the pre-processing stage to eliminate the distortion present in the image. Then, the noise-free image is given to the segmentation stage. For segmentation, in this paper, modified regularized K-means (MRKM) clustering algorithm is presented. After the segmentation process, the segmented nodule image is fed to the classification stage to categorize the nodule as benign or malignant (risk nodule). For classification, an improved convolution neural network (ICNN) is presented. The proposed ICNN is designed by modifying CNN with the integration of the adaptive tree seed optimization (ATSO) algorithm. Finally, the stage identification is carried out based on the size of the nodule and we classify the malignant nodule as S1-S4. The presented technique attained the maximum accuracy of 96.5% and performance compared with existing state-of-art methods.

BiRPN-YOLOvX: A weighted bidirectional recursive feature pyramid algorithm for lung nodule detection.

BACKGROUND: Lung cancer has the second highest cancer mortality rate in the world today. Although lung cancer screening using CT images is a common way for early lung cancer detection, accurately detecting lung nodules remains a challenged issue in clinical practice. OBJECTIVE: This study aims to develop a new weighted bidirectional recursive pyramid algorithm to address the problems of small size of lung nodules, large proportion of background region, and complex lung structures in lung nodule detection of CT images. METHODS: First, the weighted bidirectional recursive feature pyramid network (BiPRN) is proposed, which can increase the ability of network model to extract feature information and achieve multi-scale fusion information. Second, a CBAM_CSPDarknet53 structure is developed to incorporate an attention mechanism as a feature extraction module, which can aggregate both spatial information and channel information of the feature map. Third, the weighted BiRPN and CBAM_CSPDarknet53 are applied to the YOLOvX model for lung nodule detection experiments, named BiRPN-YOLOvX, where YOLOvX represents different versions of YOLO. To verify the effectiveness of our weighted BiRPN and CBAM_ CSPDarknet53 algorithm, they are fused with different models of YOLOv3, YOLOv4 and YOLOv5, and extensive experiments are carried out using the publicly available lung nodule datasets LUNA16 and LIDC-IDRI. The training set of LUNA16 contains 949 images, and the validation and testing sets each contain 118 images. There are 1987, 248 and 248 images in LIDC-IDRI's training, validation and testing sets, respectively. RESULTS: The sensitivity of lung nodule detection using BiRPN-YOLOv5 reaches 98.7% on LUNA16 and 96.2% on LIDC-IDRI, respectively. CONCLUSION: This study demonstrates that the proposed new method has potential to help improve the sensitivity of lung nodule detection in future clinical practice.

The role of radiology in addressing the challenge of lung cancer after lung transplantation.

The importance of lung cancer as a complication of lung transplantation is increasingly recognised. It may become an important survival-limiting factor in lung transplant patients as management of other complications continues to improve and utilisation of extended criteria donors grows. Radiology can play a key role in tackling this issue at multiple stages in the transplantation pathway and follow-up process. Routine chest CT as part of pre-transplant recipient assessment (and donor assessment if available) can identify suspicious lung lesions with high sensitivity and detect chronic structural lung diseases such as pulmonary fibrosis associated with an increased risk of malignancy post-transplant. Pre-transplant CT also provides a comparison for later CT studies in the assessment of nodules or masses. The potential role of regular chest CT for lung cancer screening after transplantation is less certain due to limited available evidence on its efficacy. Radiologists should be cognisant of how the causes of pulmonary nodules in lung transplant patients may differ from the general population, vary with time since transplantation and require specific recommendations for further investigation/follow-up as general guidelines are not applicable. As part of the multidisciplinary team, radiology is involved in an aggressive diagnostic and therapeutic management approach for nodular lung lesions after transplant both through follow-up imaging and image-guided tissue sampling. This review provides a comprehensive overview of available clinical data and evidence on lung cancer in lung transplant recipients, and in particular an assessment of the current and potential roles of pre- and post-transplant imaging. KEY POINTS: • Lung cancer after lung transplantation may become an increasingly important survival-limiting factor as mortality from other complications declines. • There are a number of important roles for radiology in tackling the issue which include pre-transplant CT and supporting an aggressive multidisciplinary management strategy where lung nodules are detected in transplant patients. • The introduction of routine surveillance chest CT after transplant in addition to standard clinical follow-up as a means of lung cancer screening should be considered.

Derivation and validation of a prediction model for patients with lung nodules malignancy regardless of mediastinal/hilar lymphadenopathy.

BACKGROUND: Clinical prediction models to classify lung nodules often exclude patients with mediastinal/hilar lymphadenopathy, although the presence of mediastinal/hilar lymphadenopathy does not always indicate malignancy. Herein, we developed and validated a multimodal prediction model for lung nodules in which patients with mediastinal/hilar lymphadenopathy were included. METHODS: A single-center retrospective study was conducted. We developed and validated a logistic regression model including patients with mediastinal/hilar lymphadenopathy. Discrimination of the model was assessed by area under the operating curve. Goodness of fit test was performed via the Hosmer-Lemeshow test, and a nomogram of the logistic regression model was drawn. RESULTS: There were 311 cases included in the final analysis. A logistic regression model was developed and validated. There were nine independent variables included in the model. The aera under the curve (AUC) of the validation set was 0.91 (95% confidence interval [CI]: 0.85-0.98). In the validation set with mediastinal/hilar lymphadenopathy, the AUC was 0.95 (95% CI: 0.90-0.99). The goodness-of-fit test was 0.22. CONCLUSIONS: We developed and validated a multimodal risk prediction model for lung nodules with excellent discrimination and calibration, regardless of mediastinal/hilar lymphadenopathy. This broadens the application of lung nodule prediction models. Furthermore, mediastinal/hilar lymphadenopathy added value for predicting lung nodule malignancy in clinical practice.

Malignant Nodules Detected on Lung Cancer Screening CT: Yield of Short-Term Follow-Up CT in Showing Nodule Growth.

BACKGROUND. Lung-RADS recommends 3-month follow-up for category 4A nodules and downgrading to category 2 of all category 3 or 4 nodules that are unchanged for 3 months or longer, indicating benign behavior. This guidance may be problematic considering the potential for slow-growing cancers in that lack of nodule growth, particularly at short follow-up intervals, may provide false reassurance. OBJECTIVE. The purpose of this study was to evaluate the yield of short-term follow-up CT in showing growth among malignant nodules detected on lung cancer screening CT. METHODS. This retrospective study included 76 patients (53 women, 23 men; median age, 68 years) with a positive lung cancer screening CT result (Lung-RADS category ≥ 3) between June 2015 and May 2021 with a subsequent lung cancer diagnosis and at least one follow-up CT examination at least 3 months before diagnostic or therapeutic intervention. Semiautomated software was used for linear and volumetric nodule measurements. Diameter was defined as the mean of short- and long-axis measurements. For solid nodules, growth was defined as an at least 1.5-mm increase in mean diameter or an at least 25% increase in volume; part-solid nodules, an at least 1.5-mm increase in solid-component mean diameter or an at least 25% increase in volume; and ground-glass nodules, an at least 3-mm increase in mean diameter or development of a new solid component within the nodule. RESULTS. Median time to growth was 13 months by linear and 11 months by volumetric measurement. Frequency of growth at 3 months was 5% by linear and 7% by volumetric measurement. By linear measurement, median time to growth and frequency of growth at 3 months were 13 months and 7% (solid nodules), 18 months and 6% (part-solid nodules), not reached and 0% (ground-glass nodules), not reached and 0% (category 3 nodules), 13 months and 6% (category 4A nodule)s, 6 months and 11% (category 4B nodules), and 12 months and 10% (category 4X nodules). CONCLUSION. Malignant nodules manifest growth slowly on follow-up CT, and 3-month follow-up CT has very low yield. Stability at 3-month follow-up should not instill high confidence in benignancy, and downgrading all such nodules to Lung-RADS category 2 may be problematic. CLINICAL IMPACT. This study highlights the possibility of slow-growing malignancy and associated challenges in application of Lung-RADS to management of unchanged nodules on follow-up imaging.